01Safety
30.0 / 30
What changed in the harness
Selection accuracy 100→100, token cost up 3%, unconfirmed writes 100%→100%.
Category breakdown
Where the score comes from.
Earned points across the four signals Gradable measures. Safety and Legibility are scored out of 30; Economics and Discoverability are scored out of 20.
02Legibility
29.6 / 30
03Economics
20.0 / 20
04Discoverability
19.2 / 20
Highest-impact fix
Estimated gain +1 pointMake target tools discoverable on the first call
Clarify tool names, decision boundaries, and required argument schemas so an agent can choose and construct the target call without exploratory steps.
Description evidence
Defects and rewrites.
1 defect found across the exposed tool descriptions. Suggested rewrites make purpose, inputs, boundaries, and returns easier for an agent to understand.
| Tool | Defect types | Suggested rewrite |
|---|---|---|
lookup_object |
no_return_description |
Look up a celestial object by Messier number (e.g. M31), NGC/IC designation (e.g. NGC7000), or common name (e.g. Andromeda Galaxy). Optionally compute visibility from a given location and time. Returns the object's catalog identifiers, coordinates, magnitude, and size, plus rise/set/transit times and peak altitude when latitude, longitude, and date are provided. |
Selection evidence
Confusable tool pairs.
1 pair where similar names or overlapping descriptions may send an agent toward the wrong tool.
| Tool A | Tool B | Confidence | Why they collide |
|---|---|---|---|
lookup_object |
search_objects |
low | A task like 'find M31' clearly maps to lookup_object (named object) while 'find galaxies in Andromeda' maps to search_objects (filter by criteria); the distinct input schemas (single name vs. filter criteria) make confusion unlikely except in vague phrasing like 'find bright objects in Orion' which could ambiguously suggest either tool. |
Compare the field